IBM has issued a new Cloud Service Description, the formal document that explains what Big Blue considers to be a cloud and how it’s effort will behave.
The first item on the list of additions in the email sent to customers is: “a new tier to the Service Level Agreement (SLA) which provides a 100% refund on monthly charges, should a service miss a 95 percent availability target.”
95 percent availability means 36 hours of downtime in a month, which is not what clouds are supposed to do.
Could IBM be softening up its users for more incidents like its June 2020 crash that took its entire cloud down for hours, including vital status information pages?
Don’t panic, IBM cloud user: this isn’t Big Blue in any way saying that you can expect one-and-a-half nines reliability in its cloud.
Instead it’s IBM catching up to rivals like Azure and AWS, which already have 100 percent refund tiers at 95 percent availability and have had them in place for over a year.
All also require users to apply for those refunds, which probably isn’t what you’ll really feel like doing after losing a day-and-a-half of uptime!
IBM’s new legalese also removes mobile apps from Big Blue’s definition of its cloud UI, leaving it with front-end options of “on-line portals, APIs, command line interfaces, or, where available, assisted ordering”.
There’s also a new set of rules for notifications of price or SLA changes: in last year’s document IBM offered only a 30-day warning of “any changes to this Service Description.”
This year’s model offers the more nuanced promise of “at at least 30 days' notice of any price increases or changes to this Service Description, and at least 90 days' notice for SLA changes.”
A clause on Apple software running on iOS has been excised from the document, as have references to POWER8-powered servers. Presumably IBM has upgraded its POWER-powered cloud to newer processors. ®